Skip to main content

Full text of "Electrolysis Propulsion for Spacecraft Applications"

See other formats

NASA Technical Memorandum 113157 AIAA-97-2948 

Electrolysis Propulsion for 
Spacecraft Applications 

Wim A. de Groot and Lynn A. Arrington 
NYMA, Inc., Brook Park, Ohio 

James F. McElroy 

Hamilton Standard, Windsor Locks, Connecticut 

Fred Mitlitsky, Andrew H. Weisberg, 
Preston H. Carter II, and Blake Myers 
Lawrence Livermore National Laboratory, Livermore, California 

Brian D. Reed 

Lewis Research Center, Cleveland, Ohio 

National Aeronautics and 
Space Administration 

Lewis Research Center 

October 1997 

Available from 

NASA Center for Aerospace Information National Technical Information Service 

800 Elkridge Landing Road 5287 Port Royal Road 

Lynthicum, MD 21090-2934 Springfield, VA 22100 

Price Code: A03 Price Code: A03 

Electrolysis Propulsion for Spacecraft Applications 

Wim A. de Groot* and Lynn A. Arlington 
NYMA Inc, NASA LeRC Group 
Brook Park, Ohio 

James F. McElroy 
Hamilton Standard 
Windsor Locks, Connecticut 

Fred Mitlitsky* , Andrew H. Weisberg tt , Preston H. Carter II* , and Blake Myers** 

Lawrence Livermore National Laboratory 

Livermore, California 

Brian D. Reed* 

NASA Lewis Research Center 

Cleveland, Ohio 


Electrolysis propulsion has been recognized over 
the last several decades as a viable option to 
meet many satellite and spacecraft propulsion 
requirements. This technology, however, was 
never used for in-space missions. In the same 
time frame, water based fuel cells have flown in a 
number of missions. These systems have many 
components similar to electrolysis propulsion 
systems. Recent advances in component 
technology include: lightweight tankage, water 
vapor feed electrolysis, fuel cell technology, and 
thrust chamber materials for propulsion. Taken 
together, these developments make propulsion 
and/or power using electrolysis/fuel cell 
technology very attractive as separate or 
integrated systems. A water electrolysis 
propulsion testbed was constructed and tested in 
a joint NASA/Hamilton Standard/Lawrence 
Livermore National Laboratories program to 
demonstrate these technology developments for 
propulsion. The results from these testbed 
experiments using a 1 -N thruster are presented. A 
concept to integrate a propulsion system and a 
fuel cell system into a unitized spacecraft 
propulsion and power system is outlined. 


Innovative new systems are being sought to 
improve mission performance and reduce cost. 
Electrolysis propulsion, either alone or combined 
with fuel cell power offers the potential to 
provide a synergistic power and propulsion 
system for small spacecraft. 

On-board propulsion systems must satisfy a 
variety of propulsion functions, including orbit 
insertion, attitude control, station keeping, 
repositioning, and primary propulsion for 
planetary spacecraft. There already exists a 
number of low thrust propulsion options to carry 
out these maneuvers. Cold gas propulsion is 
commonly used when propulsion requirements 
are small and where cost and system simplicity 
are decisive factors. Monopropellant hydrazine 
(N2H4) systems are generally used for orbit 
insertion of smaller satellites because of its 
higher specific impulse (Isp) compared to cold 
gas systems. However, monopropellant systems 
are more costly and complex than cold gas. 
Storable bipropellants, utilizing nitrogen 
tetroxide (NTO) as oxidizer and either 
monomethylhydrazine (MMH) or N2H4 as fuel. 

Sr. Research Engineer, Senior Member AIAA 

Research Engineer, Member AIAA 

Program Manager 

Program Manager, Member AIAA 

Space Group Scientist. Member AIAA 

Aerospace Engineer, Member AIAA 

Mechanical Engineer, Associate Fellow AIAA 

NASA TM-1 13157 

have been used extensively for orbit insertion of 
medium to large satellites and for primary 
propulsion in planetary spacecraft. These systems 
in tum are more costly and complex than 
monopropellant systems. 

A recent trend is toward the use of electric 
thruster systems for satellite on-orbit functions. 
For example, arcjets are already used for North- 
South station keeping of geostationary satellites. 
High power ion and Hall thrusters are being 
developed for orbit transfer and primary 
planetary propulsion missions. Pulsed plasma 
thrusters are poised to be flight tested for 
precision on-orbit functions on smaller satellites. 

Water electrolysis propulsion can provide higher 
performance than the established chemical 
propulsion options. At equal thrust levels, power 
requirements of water electrolysis propulsion 
(-0.17 N/kW) are greatly below those of electric 
propulsion devices (-0.08 N/kW for 2.2 kW 
arcjets, and 0.03 N/kW for 2.6 kW ion thrusters). 
These advantages become more pronounced at 
lower power levels, where efficiencies of electric 
propulsion devices are significantly reduced. In a 
water electrolysis propulsion system, water 
stored in a lightweight, low pressure tank is fed 
to an electrolyzer. The electrolyzer consumes 
electrical energy to decompose the water into 
pressurized hydrogen and oxygen. If solar energy 
is available, these devices can also serve as a 
load leveling function, storing the energy as 
hydrogen and oxygen gases. The propellant is 
clean and inexpensive, reducing costs associated 
with propellant acquisition, ground handling, 
maintenance, and launch. Water can be stored in 
compact, lightweight tanks at relatively high 
density (1.0 g/cc). Storage requirements for 
propulsion are set by one or more high impulse 
"burns", where the hydrogen and oxygen are 
stored in separate tanks, to be mixed and ignited 
inside the combustion chamber of a conventional 
rocket engine. The gaseous hydrogen/gaseous 
oxygen (GH2/GO2) propellants have performance 
measured at an Isp of over 350 s (at thrust levels 
of 0.5 to 15 N), 3 which is superior to earth 
storable chemical alternatives. The products of 
combustion are clean and free of carbon, sparing 
optics and other sensitive instruments from 
degradation. Contamination issues with water 
vapor condensation are mission dependent and 
need to be investigated. 

Neither mechanical pumps nor pressurant gas are 
required to feed a water electrolysis rocket 
system, because electrolyzers are now able to 
electrochemically "pump" water decomposition 

products from ambient pressure up to pressures 
of at least 20 MPa. The absence of a 
pressurization system simplifies the propellant 
feed significantly and eliminates components that 
must have long-term compatibility with 
propellants. For deep space missions, water is 
significantly easier to contain than the hypergolic 
Earth storables, offering stability over a 
relatively wide temperature range. A final 
advantage of the water rocket is its dual mode 
potential. For relatively high thrust applications, 
the system can be used as a bipropellant engine. 
For low thrust levels and/or small impulse bit 
requirements, cold gas oxygen can be used alone. 

The potential of the water electrolysis rocket as a 
high performance propulsion device has been 
recognized for some time. Newman discussed 
water electrolysis propulsion for reaction control 
systems (RCS) in 1965. Stechman et al. 5 
demonstrated that 500,000 N-s of total impulse 
could be obtained with a water electrolysis 
satellite propulsion system during laboratory 
tests with 20 N and 0.5 N engines. Such a 
propulsion system, however, was never accepted 
for a flight program. This was partly due to the 
decision that the improved performance was not 
sufficient to mitigate the perceived increase in 
complexity. Other disadvantages included: the 
large tankage needed for gaseous storage, the 
increased weight due to the need to pressure feed 
the electrolyzer, the limited power available for 
propellant generation, the propellant utilization 
penalty of gas dryers, and the ignition 

Recent advances in propellant storage 
technology, 6 water vapor feed electrolysis, 7 ' 8 and 
solar array performance, along with a flurry of 
research in GH 2 /G0 2 ignition (e.g. the LEAP 
program and SSTO, 9 among others) have made 
the use of electrolysis propulsion more attractive 
from a mass standpoint. In addition, there now 
exists an innovative new system which improves 
the performance of small spacecraft called the 
Unitized Regenerative Fuel Cell (URFC), an 
integrated electrolyzer and fuel cell in a single 
reversible unit. 7 This system offers the potential 
for dual use (power and propulsion) and a 
substantial weight savings over established, 
separate, propulsion and power systems in 
certain mission scenarios. A Hamilton Standard 
study 8 showed that for low-earth-orbit (LEO) 
satellites, the specific energy (energy capacity 
per weight of storage unit) of a water fuel cell 
was better than state-of-the art NiCad batteries 
and approximately equal to that of NiH batteries, 
about 15 W-hr/kg. This study did not include the 


lightweight tankage proposed in the current 
system, which would provide higher specific 
energy. Integrating the fuel cell system with an 
electrolysis propulsion system further reduces the 
combined propulsion and power system weight 
due to common components, such as gas storage 
and the electrolyzer/fuel cell. The energy density 
of such a unitized system for LEO applications 
increases an order of magnitude (-150 W-hr/kg). 
Also, the weight advantage of both stand alone 
fuel cells and unitized systems increases for 
missions with a longer energy charge-discharge 
cycles. This results from the separation of power 
and energy inside the URFC. Batteries scale 
linearly with energy storage requirement, 
whereas for URFC's, only the storage tanks scale 
with energy storage requirements. The reactor 
stack is scaled only for power. 

Perceived system complexity can be a major 
obstacle to in-flight use. The addition of an 
electrolyzer to the propulsion system slightly 
increases complexity over a gas pressurized 
system. However, the combination of a tenfold or 
more reduction in combined propulsion/power 
system mass over state-of-the-art systems and the 
cleanness of propellants can favor a more 
complex system. 

The full advantage of electrolysis propulsion is 
gained when possible synergies with other 
subsystems are realized. A schematic of such a 
proposed unitized system is shown in Fig. 1. 
Because most of the power for flight electronics 
isn't required during orbital transfer maneuvers, 
it will often be available to electrolyze water 
without adding additional capability and mass 
penalty. High performance gas storage tanks can 
provide some, if not most of the structure 
required by spacecraft that must function as stiff 
instrument platforms. A unitized propulsion and 
power system was proposed for a New 
Millennium Program spacecraft concept. For the 
system proposed, a URFC was used to replace 
the baseline batteries for energy storage. The 
modest 30% increase in electrolyzer mass was 
more than offset by the savings in battery mass 
which accounted for as much as 10% of the wet 
mass. The projected benefits of such an 
integrated system were a weight savings of over 
50% for low-earth-orbit spacecraft, increasing 
with higher energy storage needs. Missions 
analyses show that electrolysis systems also 
provide significant weight savings for 
applications which require a large number of 
impulsive burns. 

This paper will first describe recent advances in 
component technologies which may make 
electrolysis propulsion a viable candidate for a 
variety of mission scenarios. This is followed by 
a description of a testbed built at NASA LeRC in 
a cooperative program partnering Lewis 
Research Center, Hamilton Standard and 
Lawrence Livermore National Laboratories, and 
results obtained from experiments in a high 
altitude simulation chamber. 

Component Technologies 

A schematic of a water electrolysis propulsion 
system which could be used to provide all 
propulsion functions in a small satellite 
application is shown in Fig. 2. It includes a 
primary thruster for high AV maneuvers, four 
cold gas thrusters for thrust vector control during 
primary burns, and twelve cold gas thrusters for 
attitude control (ACS). This system is designed 
to replace two conventional (i.e. cold gas and 
N 2 H 4 ) systems that would be needed to perform 
the same functions in a mission utilizing state-of- 
the-art technology. Key components of the water 
electrolysis system are discussed below. They are 
the electrolyzer, gas dryers, the water and 
propellant tankage, the propellant feed system, 
and the thrusters. In addition, the technology to 
integrate propulsion and power is discussed. 


A detailed description of the water vapor feed 
electrolyzer is given in Reference 7. This 
electrolyzer is based on Hamilton Standards' 
solid polymer electrolyte (SPE®) technology. The 
electrolyzer uses this sulfonic acid proton 
exchange membrane as the sole electrolyte. The 
membrane is fashioned into electrochemical cells 
by bonding catalyst electrodes to both faces. The 
single electrolysis cell consists of a water feed 
chamber, a water permeable membrane, a 
hydrogen chamber, a SPE membrane, an oxygen 
chamber, an electrochemical hydrogen pump, 
and electrical insulators on both end plates. 
Hydrogen and oxygen are produced on either 
side of the SPE membrane with the application of 
DC power. The water feed chamber is separated 
from the hydrogen gas chamber by water 
permeable membranes which allow osmotic 
water transport into the hydrogen chamber. 
Because water is being consumed to produce 
propellants, a water gradient is established across 
the water feed barrier and more water from the 
storage tank enters the cell. An electrochemical 

NASA TM-1 13157 

hydrogen pump, drawing a few milliwatt assures 
that no hydrogen builds up in the water feed 

The reliability of the water vapor feed 
electrolysis system has been demonstrated 
previously in an accelerated test simulating 10 
years worth of propellant production for North 
South station keeping (NSSK) on a 
geosynchronous satellite. 10 Utilizing the 
electrochemical "pumping" action of the SPE 
electrolyzer, gaseous hydrogen and oxygen up to 
pressures of 2.72 MPa (20 MPa has been 
demonstrated) were produced, with subsequent 
burns consuming propellants down to 0.7 MPa 
tank pressure. SPE-based fuel cells have flown 
on seven Gemini missions," but SPE-based 
vapor feed electrolyzers have not been flight 
qualified yet. Sizing of the electrolyzer for 
selected missions depends on the systems design 
approach. Either the electrolyzer is scaled 
according to the available power and the mission 
is accomplished with the given propellant 
generation rate, or the electrolyzer is scaled 
according to the mission requirements which 
dictate the required propellant generation rate 
and therefore power. In this case, additional solar 
collectors to drive the electrolyzer are added. On 
high delta-V missions, the higher Isp of the 
hydrogen/oxygen propellants compensates for 
the additional mass of components (e.g., 
electrolyzers, gas tanks, additional solar 
collectors) that state-of-the-art chemical 
propulsion systems do not require. 

Gas Dryers 

Both the hydrogen and the oxygen leaving the 
electrolysis unit contain small quantities of water 
vapor. If not removed, this water vapor could 
condense inside the tanks and propellant lines. 
Furthermore, the presence of water vapor inside 
the propellants will reduce thruster performance. 
The installation of propellant dryers based on a 
desiccant bed is a simple solution. This would be 
a highly reliable passive component. For small 
spacecraft applications, the amount of water 
vapor will be low, so this component will be 
small with relatively low weight. The amount of 
water vapor depends on gas pressure. A 
conservative estimate is that for a 7.0 MPa 
system, approximately 2% need to be added to 
the propellant mass in order to account for the 
desiccant mass. The amount of water absorbed in 
the desiccant under these conditions is 
approximately 0.25 % of the total water wet 

Propellant Feed System 

The propellant feed system described here is 
designed for maximum simplicity. Pressurization 
of the propellants is accomplished through the 
electrolyzer. Direct feed lines from the 
electrolyzer to the tanks supply propellants. For 
highly controllable impulse bits and maximum 
combustion efficiency, regulators are needed 
between the tanks and thruster to control the 
propellant mass flow rates. For less restrictive 
needs, a blowdown system could be used to 
simplify the operation and reduce system weight 
resulting in some performance reduction. 

Over the last several years, strict micro- 
propulsion requirements have driven the 
development in valve and regulator 
technologies. This has resulted in the reduction 
of leak rates (internal leakage <10" 6 scc/h He for 
valves and < 1 scc/h for regulators, 
respectively), minimizing power requirements (< 
9 Watts), and minimizing mass (10-100 gms). 
In order to satisfy even stricter requirements, 
near term developments are focused on micro- 
electromechanical systems (MEMS) technology 
to further reduce the mass and achievable flow 
rates. The biggest obstacle with MEMS, 
however, is the leak rate, which has been greater 
than for conventionally manufactured valves, and 
the need to filter even the smallest particles. 

Water and Propellant Tankage 
Because the vapor feed electrolyzer pressurizes 
the propellant, the water supply can be stored at 
ambient pressures in thin-walled, light weight 
tanks. The storage of gaseous reactants, 
especially hydrogen, however, has always been a 
problem for on orbit applications. For missions 
in which a velocity change must be accomplished 
in a single, large AV burn, the required tank mass 
to contain the required gaseous hydrogen is high. 

If multiple burns are possible to accomplish the 
mission, filling and draining gas storage pressure 
vessels multiple times can effectively reduce the 
mass penalty of gaseous hydrogen storage. The 
propellant tanks are now sized to accommodate 
only the largest burn of the mission, the required 
mass is effectively "amortized" over the number 
of times that the tank gets refilled during the 

The figure of merit for lightweight pressure tanks 
is the performance factor, which is the burst 
pressure multiplied by the internal volume and 
divided by the tank weight (P b .V/W). Recent 

NASA TM-1 13157 

work on propellant tankage 6 ' 8 has greatly 
improved the performance factor. State of the art 
performance factors are 4 million-cm for large 
tanks (lower for smaller tanks), with a safety 
factor (maximum expected operating pressure / 
burst pressure) of 1.5. Because tanks are 
generally assumed to be pressurized in flight, this 
safety factor is conservative for tanks that are not 
pressurized when humans, launch vehicles, or 
other spacecraft are at risk. The performance 
factor is aggressive compared to commercially 
available space qualified pressure vessels which 
have a performance factor of 2 million-cm. 
However, aggressive performance factors are 
feasible using thin bladder-liners overwrapped 
with T1000 carbon fiber composite. Prototype 
bladder-lined tanks of modest size have recently 
been fabricated which achieved 4 million-cm 
using thick end domes and two heavy stainless 
steel bosses sized for automotive applications. 
Reducing the mass of the bosses and end domes 
should enable 5 million-cm tanks for large 
volumes and 4 million-cm tanks for modest 
volumes. Small tank volumes (which generally 
result in low performance factors) are readily 
contained within required structural members. 
Thus, aggressive performance factors are 
justified even for small volumes, if only the mass 
increment of turning structural members into 
pressure vessels is considered as tank weight. 
This results in a significant weight reduction as 
compared to the use of conventional tankage. 


For the current study, a 1-N GH 2 /G0 2 thruster 
was build into the testbed. This thruster consisted 
of an ignitor, an injector, a chamber, a throat, and 
a 23.3:1 area ratio nozzle. Small GH 2 /G0 2 
thrusters have been developed and tested over 
the last three decades. 13 Flight type thrusters built 
for satellite electrolysis propulsion concepts 
(thrust levels from 0.5 to 22 N) have been tested 
extensively. 4,5 ' 14 A 22-N thruster demonstrated 
over 69,000 firings with a total of 4 hours burn 
time without noticeable degradation, achieving 
an Isp of 355 s. In the same program, a 0.5-N 
thruster demonstrated over 150,000 firings and 
10 hours total burn time, with a performance of 
331 s. These tests showed that for these small 
thrusters, optimal ignition was achieved at higher 
chamber pressures (>160 kPa), driving optimal 
designs to operate at higher tank and electrolysis 

Thrusters built for potential application as the 
space station propulsion system (thrust levels 
from 110 to 220 N) have also been tested 

extensively. 15 ' 16 These non-optimized thrusters 
have achieved Isp's up to 360 s at stochiometric 
mixture ratio. Most recently, 2200-N, GH 2 /G0 2 
thrusters were developed for the X-33, the 
technology demonstrator vehicle for the 
Reusable Launch Vehicle. 9 

In all of the past work, fuel-film cooling was 
used for thermal and oxidation protection of 
thruster walls. The presence of such a fuel-film 
reduced thruster performance. In order to 
maximize thruster performance in the highly 
oxidizing combustion environment of a 
stochiometric GH 2 /G0 2 thruster, advanced 
thruster materials, such as iridium-coated 
rhenium (Ir/Re) may be needed. This material 
provides a 700 K increase in operating 
temperature over the best state-of-the-art 
chamber material. Ir/Re rockets have allowed the 
virtual elimination of fuel-film cooling for 
storable bipropellants, resulting in greatly 
improved performance. 17 As the result of an 
intensive development program, these thrusters 
are close to being commercially available. For 
stochiometric GH 2 /G0 2 , Ir/Re with an additional 
oxide coating for increased oxidation-resistance 
may be a better option. Several 22-N, oxide- 
coated Ir/Re thrusters have been tested on 
GH 2 /G0 2 up to a mixture ratio of 17. 18 

Leveraging the results of advanced thruster 
materials research and redesigning thrusters to 
operate with radiative cooling alone, can increase 
specific impulse by a significant margin 
(projected Isp > 380 s) while at the same time 
operating in an oxidizing environment. The 
additional performance that could be obtained 
from GH 2 /G0 2 systems is higher than from 
storable propellant systems using the same 

One major difference between GH 2 /G0 2 and 
established chemical thrusters is the need for an 
ignition source. Incorporation of an ignition 
source may increase complexity or power 
requirements and may not meet the stringent 
pulsing requirements of some low thrust rockets. 
Spark ignition has been used extensively in 
previous GH 2 /G0 2 thruster programs and is the 
baseline for the X-33 thruster. Alternative 
ignition sources, including laser, resonance, and 
catalytic ignition have also been investigated for 
GH 2 /G0 2 . 19 Ignition systems are being 
investigated under technology programs for 
upgrade of the Shuttle Orbiter RCS and manned 
lunar/Mars spacecraft, both of which will 
probably use oxygen/hydrocarbon propellants. 

NASA TM-1 13157 

Integrated Propulsion and Power 
Missions amenable to electrolysis propulsion can 
gain from having both the electrolyzer and the 
batteries replaced with a URFC. 7 In this case, the 
weight of the unitized system is shared by the 
power and propulsion system thus providing a 
savings over conventional systems. Recent 
results have demonstrated that URFCs are 
capable of many energy storage cycles without 
significant degradation. 6 Results from recent 
accelerated cycle testing are shown in Fig. 3 
along with a description of the single cell URFC 
cycle test conditions. More than 2010 alternate 
cycles of fuel cell (FC) and electrolyzer (EC) 
operation were accomplished at four different 
power levels. Critical system parameters did not 
change over the course of the test, indicating that 
life and also the system operated over a wide 

These results indicate that URFCs should be able 
to power satellites through many thousands of 
eclipse periods. Unlike battery power systems 
which require shallow depth of discharge to 
achieve long cycle life, URFC energy storage 
systems should be capable of deep discharges 
throughout their entire service life. 

Table I gives a summary of the status of the 
different technologies. All technologies have 
demonstrated performance at NASA's 
technology readiness level 4 or higher. 

polysulfone cell frames. The unit was designed to 
operate at pressures as high as 1 MPa. With the 
water tower filled up to 15 cm, the total impulse 
of this system was estimated to be 1000 N-s if an 
Isp of 330 s is assumed. 

Hydrogen, generated inside the electrolysis cell 
percolated to the top of the tower. A compression 
fitting installed in the tower wall connected to a 
3.18-mm diameter propellant line, which 
supplied hydrogen to a 300-cc storage tank, rated 
for 20 MPa. Oxygen generated inside the 
electrolysis cell accumulated inside the base. 
Another fitting in the side of the base connected 
to a 3.18-mm diameter propellant line, supplying 
oxygen to a 150-cc storage tank. The tanks were 
designed to assure nearly equal pressures based 
on the decomposition. 

Solenoid valves installed between the electrolysis 
unit and the storage tanks were opened during the 
electrolysis cycle and then closed during thruster 
firing. The valve closing prevented water from 
being drawn from the electrolysis tower into the 
propellant lines by sudden depressurization 
following ignition. This valve would be 
eliminated in a true flight design by the use of a 
zero gravity compatible water vapor feed 
electrolyzer. Nitrogen purge lines between the 
tanks and the electrolysis unit allowed the 
propellants to be purged, exhausting through the 
rocket nozzle. This feature was only required in 
ground testing. 

Electrolysis Propulsion Breadboard Tests 

As a proof of concept, a complete electrolysis 
propulsion system was assembled. A schematic 
of the electrolysis breadboard system is shown in 
Fig. 4. For simplicity, power was obtained from a 
35 V power supply, to simulate the small 
spacecraft bus. The maximum available power 
was 700 W. The system was designed to operate 
in blowdown mode (i.e. no regulators were used). 
A description of the system components follows. 

In a flight qualified system, the electrolyzer used 
would be a zero gravity compatible water vapor 
feed electrolyzer. The electrolysis unit used in 
the current experiments, however, was not a 
flight-type unit, but was a commercial, 
percolating, cathode gravity liquid feed 
electrolyzer provided by Hamilton Standard. 
This unit consisted of a 5-cm diameter, 20-cm 
high, plexiglass water tower on a 12.5-cm square, 
5- cm high base. The electrolysis cell was housed 
in the base of the unit and was a 45.2-cm 2 , 
platinized Nation 117 membrane with 

Sonic Venturis installed inside the propellant 
lines downstream of the storage tanks fixed the 
propellant mass flow rates to the thruster. The 
Venturis were designed for specific mass flow 
rates at inlet pressures of 0.68 MPa to achieve a 
stochiometric mixture ratio of eight. However, 
the Venturis were calibrated over a range of inlet 
pressures. The mass flow rates, and thus the 
chamber pressure, decreased during a blowdown 
test, as the inlet pressures vary from 1.0 to 0.5 
MPa. Calibration data assured that the Venturis 
were choked at all points during blowdown tests 
for these operating conditions. 

Opening of thruster valves, installed downstream 
of the Venturis, caused the Venturis to choke, 
controlling hydrogen and oxygen mass flows to 
the injector. The injector available for these tests 
was optimized for a 20-N thruster. As a result, 
the injector did not provide optimum 
performance for the current tests, but was good 
enough for the purpose of this study. The 
oxygen was injected into a center annulus, where 
it was excited by a spark ignition system. Six 

NASA TM-1 13157 

small slots on the back of a hydrogen splitter ring 
provided radial injection of the "igniter 
hydrogen", while six elements canted inward 
provided hydrogen injection further downstream. 
No film cooling was employed. A 5-cm long 
water-cooled adapter, with a stainless steel 
boundary layer trip ring, provided additional 
mixing and was used to mount the chamber to the 

Two chambers were tested with the injector. A 
copper heat-sink chamber was used for checkout 
of the system, and an Ir/Re chamber was then 
installed for the majority of testing. The Ir/Re 
chamber, designed for 1 -N thrust, consisted of a 
8.98-mm diameter chamber and a 2.41-mm 
diameter throat. The nozzle expansion ratio was 
23.3. It had previously undergone life testing and 
had an accumulated test time of 11.5 hours at a 
mixture ratio of 5. The copper chamber had a 
similar diameter chamber, a 2.43 mm diameter 
throat, but a slightly shorter chamber and 
different converging section. 

Thermocouples and pressure transducers at 
selected locations near the electrolysis unit, the 
storage tanks, and the thruster, were used to 
monitor temperature and pressure conditions. 
Particle filters downstream of the storage tanks 
removed particles larger than 15 micron to 
protect valve seats and injector ports. Relief 
valves rated at 1 .0 MPa near the electrolysis unit 
protected the unit from over pressurization in the 
case of component malfunction. 

Experimental Approach 

The breadboard system was installed and tested 
inside the high altitude simulation test facility 
described in Reference 20. Figure 5 is a 
photograph of the test configuration. Ambient 
pressure in the altitude chamber during the test 
was maintained at approximately 1 kPa using a 
two-stage ejector. Key data were obtained during 
the testing of the breadboard propulsion system, 
both during the propellant generation as well as 
during the hot-fire test with the thruster. 

Key parameters, measured and recorded during 
the electrolysis fill cycle, were tank pressures and 
temperatures, electrolysis pressure and 
temperature, ambient pressure and temperature, 
and electrolysis current and voltage. The last two 
variables were determined by the available 
power. Parameters recorded during hot-fire tests 
were the pressures and wall temperatures in the 
combustion chamber, the pressure drop in the 
tanks in 0. 1 s increments, venturi inlet pressures 
and temperatures, and ambient pressure and 

temperature. All data were recorded with a stand- 
alone data acquisition system and stored in a 
personal computer. 

In addition to the measured parameters, some 
additional quantities were calculated. Propellant 
flow rates could be calculated from the venturi 
inlet pressures, temperature, and calibration. 
Both the theoretical and experimental 
characteristic velocity C*, which is a measure of 
combustion efficiency, could be determined with 
standard methods and using the CEC (chemical 
equilibrium code) 21 for the given propellant 
mixture ratio. The C* efficiency, defined as the 
ratio of experimental versus theoretical 
characteristic velocity, was also determined. 

In preparation for a series of tests, all air from 
the electrolysis unit, storage tanks, and propellant 
lines was evacuated by means of opening the 
valves to the high altitude environment. After 
propellant system evacuation, the thruster valves 
were closed, the supply valves opened, and 
power was supplied to the electrolysis unit. 
Hydrogen and oxygen were generated and the 
storage tanks were filled to a predetermined 
pressure of around 1 MPa. Different power levels 
were applied at a number of electrolysis cycles in 
order to establish conversion efficiency 
variations for varying propellant generation rates. 
The duration of the propellant fill was between 
twenty minutes and several hours, depending on 
the power level. Data were taken at five minute 

A rocket firing followed each tank fill. Thruster- 
valve opening and spark ignition initiated 
combustion. The lead time between the spark 
ignition and the thruster valves opening was pre- 
set. For most of the tests reported in this paper, 
spark ignition and thruster valve opening 
occurred simultaneously. Combustion chamber 
pressures decreased during a typical blowdown 
test from 190 to 138 kPa. This range was 
selected as it bounds the design point of the 
chamber (170 kPa). A typical test duration was 
3-4 s, which was limited by the volume of the 
tanks and the maximum pressure allowed with 
the present electrolysis system. Hot-fire tests 
were terminated after the chamber pressure 
dropped below a pre-set value, which was 
selected to provide an acceptable combustion 
efficiency during this blowdown test. During hot- 
fire tests, data were taken at 100 ms intervals. 

Test conditions varied during a sequence of hot- 
fire tests as the result of changing system 
conditions. The volume occupied by hydrogen 

NASA TM-1 13157 

consisted of the storage tank, propellant line and 
head space inside the water tower of the 
electrolysis unit. The volume occupied by the 
oxygen consisted of the storage tank and the 
propellant lines. During the initial tank fill, from 
high altitude ambient up to 1 MPa, the pressure 
inside the hydrogen tank increased more slowly 
than inside the oxygen tanks due to the 
additional head space. The pressure in both tanks 
remained steady after closing the supply valves. 
This caused the mixture ratio of the first hot-fire 
in a test sequence to be oxygen rich (O/F -9.2). 

Because high pressure hydrogen was trapped 
inside the electrolysis tower, the hydrogen tank 
experienced an increase in pressure each time the 
supply valve opened. As a result, the pressure 
inside the hydrogen tanks was higher than in the 
oxygen tanks during subsequent tests, causing a 
slightly hydrogen rich mixture ratio (O/F -7.6). 
About 8 test sequences were required to reach 
marginal equilibrium conditions, because the 
space inside the water tower changed as a result 
of water consumption. Even though conditions 
changed slightly during continuing testing, 
chamber pressure, O/F ratios, and characteristic 
velocities did not change noticeably. Throughout 
the full course of testing, the thruster performed 

Electrolysis System Performance 
Key parameters during an electrolysis tank fill 
were the supplied voltage, the current through the 
cells, the pressure build-up inside the oxygen and 
hydrogen storage tanks, the electrolysis unit 
temperature, and the rate of propellant generation 
(measured in total fill time to an oxygen tank 
pressure of 1 MPa). The electrolysis voltage 
provides a measure of cell conversion efficiency 
This efficiency decreases with increasing cell 
current and electrolysis pressure, and decreases 
slightly with cell temperature. Cell voltage 
ranged from 1.47 V at 1 kPa and 1A to 1.81 V at 
1 MPa and 10 A. Electrolysis tests were 
performed at a variety of different cell currents, 
from 2 to 10 A. The current was kept at a 
constant value during each test. The increasing 
pressure inside the electrolysis unit during each 
test caused the cell voltage to gradually increase, 
requiring a slightly higher power for conversion 
than at lower operating pressures. The constant 
current assured a constant propellant generation 

Fig. 6a, b, and c show the electrolysis power 
required to maintain constant propellant 
generation rate with increasing pressure, for 
oxygen generation rates of 7.5, 18.7, and 37.5 

seem, respectively, which correspond to 2, 5, and 
10 A cell current. These cell currents translate to 
approximately 3, 8 and 16 W available power, 
typical for small spacecraft. The horizontal axis 
displays the pressure in the oxygen storage 
volume. The range displayed is from 0.6 to 1.0 
MPa, approximately the pressure range when 
cycling between electrolysis charge and hot-fire 
discharge. The vertical axis shows the power 
required. Fig. 6 shows that the required power 
increases, as expected, with increasing storage 
pressure, and that this increase is larger for 
higher generation rates. These experiments 
showed that the electrolysis conversion 
efficiency decreased gradually with increasing 
pressure, as expected by theory, due to the 
energy required for gas compression and to the 
internal hardware configuration. This pushes 
design tradeoffs of an electrolysis propulsion 
system toward lower maximum electrolysis 
pressure in order to maximize efficiency. The 
stepwise increase in Fig. 6a is due to the 
characteristics of the data acquisition equipment. 

Increasing the input power leads to an increase in 
propellant generation rate. Fig. 7a shows the 
average power required as a function of oxygen 
generation rate. The vertical axis of this figure is 
taken as the average power required between 0.6 
and 1.0 MPa oxygen storage pressure. The figure 
shows an approximately linear relationship 
between input power and generation rate, with a 
value of 0.46 W/(sccm oxygen). 

As was shown in Fig. 6, a difference exists 
between power required at 0.6 MPa, and 1.0 
MPa. This is displayed in Fig. 7b, which shows 
that the absolute difference between power levels 
required at 0.6 MPa and 1 .0 MPa increases with 
increasing propellant generation rates. The 
average electrolyzer efficiency is defined as the 
minimum power theoretically required for water 
electrolysis divided by the actual power used. 
The remaining power is rejected as heat. Typical 
efficiency values for electrolysis are between 85 
and 90%. 

Rocket Testing 

As noted previously, initial rocket test sequences 
were executed with a copper heat-sink chamber. 
Temperature, pressure, and propellant mixture 
ratio data were obtained to verify that test 
conditions remained within their expected 
ranges. Typical copper combustion chamber 
pressures are shown in Fig. 8 as a function of 
time. At the initiation of testing, the thruster 
valves opened, the spark igniter was turned on, 
and the chamber pressure increased as the result 


of gas inflow. The first pressure increase was 
detected after 100 ms. The likely cause of this 
delay in measured pressure rise was slow 
dynamics in the pressure sensing port inside the 
injector. During the next 100 ms, the pressure 
increased to approximately 69 kPa. A slight 
(100-150 ms) hesitation was detected before 
ignition occurred. The pressure rise through 
ignition was not smooth. Such a "step" was 
undesirable for performance reasons. A similar 
step was found in testing at Marquardt with a 
0.45 N thruster. 2 Further development under that 
program succeeded in eliminating this 
undesirable phenomenon by a redesign of the 
ignitor. The step was present during both the 
copper chamber tests, and the Ir/Re thruster tests, 
suggesting that it was caused by the 
injector/igniter design which was not optimized 
for these laboratory experiments. 

Ignition occurred at approximately 250 ms after 
test initiation, after which the pressure increased 
sharply until hot test equilibrium conditions were 
reached. Subsequently, the chamber pressure 
decreased as the propellant supply pressures 
decrease and less mass flowed into the chamber. 

described, the first hot-fire test experienced a 
high O/F ratio of 9.3 due to the higher oxygen 
tank pressure. This caused the oxygen mass flow 
rate to be greater than stoichiometric. Therefore, 
during the test, the O/F ratio dropped slightly. 
Subsequent tests showed lower O/F values, with 
an approximate equilibrium reached at an O/F of 
7.5. The variation in mixture ratio was caused by 
the particular geometry used in the bench test, 
where the hydrogen storage volume was more 
than twice the oxygen storage volume. A 
configuration designed for optimum performance 
is not expected to show this large variation, but is 
expected to operate at a nearly constant O/F of 

Fig. 9c shows the C* efficiency. It shows that the 
maximum C* efficiency was obtained after 
approximately 1.2 s. This indicated that a 
significant fraction of the propellant mass was 
expelled before optimum conditions were 
reached. The maximum C* efficiency was 
approximately 0.79. This level of performance 
was expected as the result of the non-optimized 
design of injector, water-cooled adapter section, 
and chamber. 

Fig. 9a show typical combustion chamber 
pressures during the Ir/Re thruster tests. All of 
the hot-fire tests show the same step in 
combustion chamber pressure increase that was 
shown in Fig. 8. Again this was attributed to the 
fact that the ignition was not optimized. Such a 
step should not present an issue in a flight type 
system. As a result of slightly different chamber 
dimensions in the Ir/Re thruster, as compared to 
the copper chamber, the cold gas pressure 
buildup reached a higher pre-ignition equilibrium 
level, -78 vs. -68 kPa; and at a later stage, -400 
ms vs. -200 ms. Ignition always occurred, with 
delays varying from 50 to 150 ms. The ignition 
delay is shown in Fig. 9a. Pressure rise after 
ignition was slow. A maximum pressure between 
173 kPa and 190 kPa, depending on mixture 
ratio, was reached -Is after test initiation. After 
that, the chamber pressure gradually dropped as 
the result of decreasing propellant supply 
pressures and thus mass flow rates. The hot-fire, 
low-pressure abort limit for this specific series of 
tests was set to 136 kPa, which ended the tests. 
The abort limit was selected to provide an 
acceptable combustion efficiency during 
blowdown tests. This was corroborated by 
alternate test series done with abort limits of 68 
kPa and 34 kPa. 

Fig. 9b shows the propellant mixture ratio (O/F 
ratio) during the series of tests. As previously 

The thruster was designed for optimum 
performance at 170 kPa chamber pressure. This 
was reached at approximately 1.5 s, which is 
indeed where the maximum combustion 
efficiency is obtained. After this maximum, the 
combustion efficiency decreases as the chamber 
pressure decreases and the conditions move away 
from optimum. External chamber wall 
temperatures did not exceed 1 800 °F. 


Electrolysis propulsion has been recognized as 
an attractive option for satellite and spacecraft 
over the decades, but has not yet been used for 
in-space missions. Recent advances in water 
vapor feed electrolysis, propellant tankage, 
thruster chamber materials, and fuel cell 
technology warrants renewed consideration for 
the electrolysis propulsion option. An electrolysis 
propulsion system would generate GH2/GO2 
propellants, without the need for a pressurization 
system, pumps, or compressors. The gaseous 
propellant tanks can be sized for the largest bum 
required for the mission, with the bulk of the 
propellant stored as water until needed. 

Electrolysis propulsion would provide higher 
performance than the established chemical 
propulsion options and at the same thrust levels. 

NASA TM-1 13157 

Furthermore, the water propellant greatly 
simplifies ground loading and servicing 
requirements and eliminates many of the 
handling compatibility concerns of toxic earth 
storable propellants. The real attractiveness of 
electrolysis propulsion might be the ability to 
serve in the dual role of propulsion and power 
generation. The tankage in this unitized system 
can also provide some, if not most of the 
structure required by spacecraft that must 
function as a stiff instrument platform. A unitized 
electrolysis/fuel cell unit would provide high 
performance propellants for propulsion and 
generate power. This might be a critical function 
for deep planetary missions, where solar power 
will become more scarce as the mission proceeds 
and where load leveling can be an important 


'Wilson, A., Jane's Space Directory. 12 lh Ed. 
1996-1997, Jane's Information Group Ltd., 
Sentinel House, Surrey, England, UK, 1994, 

2 Myers, R.M., Oleson, S.R., Curran, F.M., and 
Schneider, S. J., "Chemical and Electrical 
Propulsion Options for Small Satellites, " 
Proceedings of the 8 th AIAA Utah State 
University Conference on Small Satellites, Aug. 
29 -Sept. 1, 1994. 

Sutherland, G. S., and Maes, M. E.: "A 
Review of Microrocket Technology: 10' to 1 Ibf 
Thrust, " J. Spacecraft and Rockets, Vol. 3, No. 
8, August 1966. 

The state of the technology for the components 
of a flight-type system were discussed and a 
propulsion breadboard system was assembled. A 
series of cycles with alternate propellant 
generation by means of water electrolysis and 
subsequent hot-fire thruster tests was 
demonstrated on this breadboard system. 
Hydrogen and oxygen produced during the 
electrolysis process were stored inside small, 
high pressure tanks. The thruster used was a high 
temperature, oxidation resistant, rhenium- 
iridium, 1 N chamber, attached to a workhorse 
injector by means of a water-cooled adapter 

Oxygen to hydrogen mixture ratios varied 
between 7.5 and 9.5, with highest C* efficiency 
at the lowest mixture ratio as expected. The 
proof-of-concept test bed that was not designed 
and optimized for performance had a maximum 
C* efficiency of 79%. Optimization of thruster 
design will generate significantly better 
performance than those of state of the art 

The tests described in this paper showed the 
fundamental feasibility of the unitized propulsion 
and power concept. URFC cycle tests and 
electrolysis propulsion tests demonstrated that 
the system worked as anticipated. A fully 
functional unitized propulsion and power system 
featuring a water vapor feed URFC is needed to 
demonstrate the great advancements that can be 
made using this technology. 

Newman, D. P., "Water electrolysis reaction 
control system, " 7 th Liquid Propulsion 
Symposium, Chemical Propulsion Information 
Agency Publ. 72, ppl05-l 14, Oct. 1965. 

5 Stechman, R.C., Campbell, J.G. "Water 
Electrolysis Satellite Propulsion System, " The 
Marquardt Company, Technical Report AFRPL- 
TR-72-132, January, 1973. 

& Mitlitsky, F., Myers, B., and Weisberg, A.H., 
"Lightweight pressure vessels and unitized 
regenerative fuel cells, " 1996 Fuel Cell Seminar, 
November 17-20, 1996, Orlando, FL; UCRL-JC- 
125220 and UCRL-MI- 125220. 

7 McElroy, J.F., "Unitized regenerative fuel 
cell storage system for aircraft and orbital 
applications, " UTC Hamilton Standard div., 
Rept. BD94-02, March 1994. 

"Mitlitsky, F., de Groot, W.A., Butler, L., and 
McElroy, J.F., "Integrated Modular Propulsion 
and Regenerative Electro-Energy Storage 
System (IMPRESS) for Small Satellites, " 10th 
annual AIAA/USU Conference on Small 
Satellites, September 16-19, 1996, Logan, UT 

*Fanciullo, T.J., and Judd, D.C., "Long Life 
Reaction Control System Design, " AIAA 
Aerospace Design Conference, AIAA Paper 92- 
0964, Irvine, CA, February 16-19, 1993. 

NASA TM-1 13157 


Campbell, J.G., and Stechman, R.C., "System 
Testing, Water Electrolysis Propulsion, " 
AFRPL-TR-74-72, The Marquardt Co., Nov. 

" Jane's Spaceflight Directory, 1987 Ed., 
Jane's Publ. Co. Ltd, London, New York, pp. 65- 
67, 1987. 

12 MOOG Space Products Division, Miniature 
Latching Solenoid Valve Data Sheets. 1996 

20 Arrington, L.A. and Schneider, S.J., "Low 
Thrust Rocket Test Facility," AIAA Paper 90- 
2503, Orlando, FL, 1990. 

21 Gordon, S., and McBride, B., "Computer 
Program for Calculation of Complex Chemical 
Equilibrium Composition, Rocket Performance, 
Incident and Reflected Shocks, and Chapman- 
Jouget Detonations," NASA SP-273, March 

13 Reed, B. D. and Schneider, S. J.: 
"Hydrogen/Oxygen Auxiliary Propulsion 
Technology, " AIAA Paper 91-3440, NASA TM- 
105249, September 1991. 

14 Rollbuhler, R. J.: "Experimental 

Performance of a Water Electrolysis Rocket, " 
NASA TMX- 1737, 1968. 

15 Richter, G. P. and Price, H. G., "Proven, 
Long-Life Hydrogen/Oxygen Thrust Chambers 
for Space Station Propulsion, " JANNAF 
Propulsion Meeting, New Orleans, Aug. 1986. 
See also NASA TM-88822. 

16 Iacabucci, R. S., et al.: "Space Station 

Technology Summary," 1989 JANNAF 

Propulsion Meeting, Vol. I, CPIA Publ. 515, 

,7 Schoenman, L.: "4000 °F for Low Thrust 
Rocket Engines, " AIAA Paper 93-2406, June 

Table I: Demonstrated Technology Readiness 




SPE Fuel Cell 

Level 9 

Gemini & Biosat. 


Level 6 

Air Force Program 



Level 4 

2010 Laboratory 


Gas Dryers 

Level 4 

JSC Program 

GH 2 /G0 2 




• Valves 

Level 8-9 

• Combustion 

Level 6 


• Ignition 

Level 4 


Level 4 

Solar Rechargeable 


Aircraft & DOE & 



18 Reed, B. D., "Long-Life Testing of Oxide- 
Coated Iridium/Rhenium Rockets," 31 st Joint 
Propulsion Conference, AIAA Paper 95-2401, 
June, 1995. 

19 Reed, B.D. and Schneider, S.J., 
"Hydrogen/Oxygen Auxiliary Propulsion 
Technology," NASA TM- 105249, AIAA Paper 
91-3440, presented at the Conference on 
Advanced Space Exploration Initiative 
Technologies, Cleveland, OH, September 4-6, 

NASA TM-1 13157 




trv Fr«m« 

in Frame 

Figure 1: Integrated Modular Propulsion and Regenerative Electro-Energy Storage System (IMPRESS) 

=. s 

i 02 Storage tank 

Vi V 

f Water Tank | 

V / 1 , J ' 

M- 1 



/ Regulaby 
f - Pilch + Rolf 

\! ! H2 Storage tank] r~r 

'"' © g Latch^l 

» + Roll i 
^ (E) 

Figure 2: Schematic of Dual-Mode Electrolysis Propulsion System 

NASA TM-1 13157 




. I * * * 
w >< M 

i i • • • 



♦ EC @ 344 mA/cm2 
"» EC @ 258 mA/cm2 
"* EC @ 172 mA/cm2 
"* EC @ 86 mA/cm2 

Active Area 46 cm2 
Nation 117, E-5™ catalyst 
Inlet Temperature 322 K 
0.38 MPaH2/H20 Vapor 
0.45 MPa 02 / H20 Vapor 
300 sec FC ©108 mA/cm2 
300 sec EC 0258 mA/cm2 

FC @ 43 mA/cm2 

• FC @ 108 mA/cm2 
-° FC @ 172 mA/cm2 

A FC @ 237 mA/cm2 

H — ' — <- 



















^ 0.5 






Cycle Number (600 sec / cycle) 

Figure 3: Measured Performance as a Function of Cycle Number for a Single Cell URFC 

lo Lxisting Regulated 
Nitrogen Purge Supply 







High temperature 
Oxidation Resistant 
1.0 N Thruster 


Solenoid Valve 
Relief Valve 
Check Valve 
Flow Venturi 




Particle Filter 
Propellant Tank 
Pressure Transducer 

Figure 4: Water Electrolysis Propulsion System Breadboard Schematic 

NASA TM-1 13157 


Figure 5: Photograph of Electrolysis Propulsion 
Breadboard Installed Inside High Altitude 



u A 


_b £& 









r A 







A Oxygen Generation 
-£ Rate: 1&7 seem 

i i i i i 


0.6 0.7 0.8 0.9 1.0 
Oxygen Pressure (MPa) 












/Y W?Y\ 
f WWN\ 


Oxygen Generation 
Rate: 7.5 seem 

0.6 0.7 0.8 0.9 1.0 
Oxygen Pressure (MPa) 







1 18.00 





■1-7 on 






Oxygen Generation 
Rate: 37.5 seem 

i i i i 

0.6 0.7 0.8 0.9 1.0 
Oxygen Pressure (MPa) 

Figure 6: Average Electric Power Required to 
Maintain Constant Propellant Generation Rate at 
Increasing Pressures. O2 Generation Rates: 
a) 7.5 seem; b) 18.7 seem; c) 37.5 seem. 
























Power Required: 
0.458 W/sccm 2 

I I I 

^ 0.20 

5 10 15 20 25 30 35 40 
Oxygen Generation Rate (seem) 




-b tf 

AP (W) - — ___ I 

1 0.10 

_(P (n (1 .0 MPa)-P ln (0.6 MPa)) / 



- 0.05 




l I I I l I 

Q 5 10 15 20 25 30 35 40 

Oxygen Generation Rate (seem) 

Figure 7: Electrolysis Power Characteristics as a 
Function of Propellant Generation Rate: a) 
Average Power Required; b) Difference in Power 
Required at 0.6 and 1 .0 MPa Pressure. 

IS" 0.10 

-O- D439_88 
-O- 0439 90 
-A- D439.92 
-*v- D439.94 
-0- D439 96 
O D439_98 
-O- D439_100 




Q. 0.05 


0.0 0.5 1.0 1.5 2.0 2.5 3.0 

time (sec) 

Figure 8: Copper Combustion Chamber 
Pressures for a Series of Hot-Fire Tests. 



£ 0.15 - 


£ 0.10 





0.0 0.5 1.0 1.5 2.0 
time (sec) 










0.0 0.5 1.0 1.5 2.0 

time (sec) 


>. 0.6 




£ 0-4 




- ^ 

i i i i 

0.0 0.5 1.0 1.5 2.0 

time (sec) 

Figure 9: Ir/Re Thruster Test Results: a) 
Combustion Chamber Pressure; b) Propellant 
Mixture Ratio; c) C* Efficiency. 

NASA TM-1 13157 



Form Approved 
OMB No. 0704-0188 

Public reporting burden for this collection of information is estimated to average 1 hour per response, including the time for reviewing instructions, searching existing data sources, 
gathering and maintaining the data needed, and completing and reviewing the collection ot information. Send comments regarding this burden estimate or any other aspect of this 
collection of information, including suggestions tor reducing this burden, to Washington Headquarters Services, Directorate for Information Operations and Reports, 1215 Jefferson 
Davis Highway, Suite 1204, Arlington, VA 22202-4302, and to the Office of Management and Budget, Paperwork Reduction Project (0704-0188), Washington, DC 20503. 

1. AGENCY USE ONLY {Leave blank) 


October 1997 


Technical Memorandum 


Electrolysis Propulsion for Spacecraft Applications 


Wim A. de Groot, Lynn A. Arlington, James F. McElroy, Fred Mitlitsky, 
Andrew H. Weisberg, Preston H. Carter U, Blake Myers, and Brian D. Reed 




National Aeronautics and Space Administration 
Lewis Research Center 
Cleveland, Ohio 44135-3191 


E- 10907 


National Aeronautics and Space Administration 
Washington, DC 20546-0001 


NASA TM- 113157 


Prepared for the 33rd Joint Propulsion Conference and Exhibit cosponsored by AIAA, ASME, SAE, and ASEE, Seattle, Washington, July 6-9, 
1997. Wim A. de Groot and Lynn A. Arlington, NYMA, Inc., 2001 Aerospace Parkway, Brook Park, Ohio 44142 (work funded by NASA 
Contract NAS3-27186); James F. McElroy, Hamilton Standard, Windsor Locks, Connecticut; Fred Mitlitsky, Andrew H. Weisberg, Preston H. 
Carter II, and Blake Myers, Lawrence Livermore National Laboratory, Livermore, California; Brian D. Reed NASA Lewis Research Center. 
Responsible person, Wim A. de Groot, organization code 5400, (216) 977-7485. 


Subject Category: 72 

Distribution: Nonstandard 

This publication is available from the NASA Center for AeroSpace Information, (301) 621-0390 


13. ABSTRACT (Maximum 200 word*) 

Electrolysis propulsion has been recognized over the last several decades as a viable option to meet many satellite and 
spacecraft propulsion requirements. This technology, however, was never used for in-space missions. In the same time 
frame, water based fuel cells have flown in a number of missions. These systems have many components similar to 
electrolysis propulsion systems. Recent advances in component technology include: lightweight tankage, water vapor feed 
electrolysis, fuel cell technology, and thrust chamber materials for propulsion. Taken together, these developments make 
propulsion and/or power using electrolysis/fuel cell technology very attractive as separate or integrated systems. A water 
electrolysis propulsion testbed was constructed and tested in a joint NASA/Hamilton Standard/Lawrence Livermore 
National Laboratories program to demonstrate these technology developments for propulsion. The results from these 
testbed experiments using a 1-N thruster are presented. A concept to integrate a propulsion system and a fuel cell system 
into a unitized spacecraft propulsion and power system is outlined. 


Electrolysis; Propulsion; Satellite 












NSN 7540-01-280-5500 

Standard Form 298 (Rev. 2-89) 
Prescribed by ANSI Std. Z39-18